Simultaneous NuSTAR/Chandra observations of the Bursting Pulsar GRO J1744-28 during its third reactivation

TL;DR

This study presents simultaneous Chandra and NuSTAR observations of the Bursting Pulsar GRO J1744-28 during its third reactivation, revealing spectral features, burst behaviors, and magnetic field estimates during active outburst.

Contribution

First simultaneous high-energy observations of GRO J1744-28 during reactivation, providing detailed spectral analysis and insights into accretion disk structure and magnetic field.

Findings

Detected up to 60 keV with Eddington flux levels

Identified spectral features including Fe lines and a 10 keV feature

Estimated magnetic field strength around 9x10^10 G

Abstract

We report on a 10 ks simultaneous Chandra/HETG-NuSTAR observation of the Bursting Pulsar, GRO J1744-28, during its third detected outburst since discovery and after nearly 18 years of quiescence. The source is detected up to 60 keV with an Eddington persistent flux level. Seven bursts, followed by dips, are seen with Chandra, three of which are also detected with NuSTAR. Timing analysis reveals a slight increase in the persistent emission pulsed fraction with energy (from 10% to 15%) up to 10 keV, above which it remains constant. The 0.5-70 keV spectra of the persistent and dip emission are the same within errors, and well described by a blackbody (BB), a power-law with an exponential rolloff, a 10 keV feature, and a 6.7 keV emission feature, all modified by neutral absorption. Assuming that the BB emission originates in an accretion disc, we estimate its inner (magnetospheric) radius to be about 4x10^7 cm, which translates to a surface dipole field B~9x10^10 G. The Chandra/HETG spectrum resolves the 6.7 keV feature into (quasi-)neutral and highly ionized Fe XXV and Fe XXVI emission lines. XSTAR modeling shows these lines to also emanate from a truncated accretion disk. The burst spectra, with a peak flux more than an order of magnitude higher than Eddington, are well fit with a power-law with an exponential rolloff and a 10~keV feature, with similar fit values compared to the persistent and dip spectra. The burst spectra lack a thermal component and any Fe features. Anisotropic (beamed) burst emission would explain both the lack of the BB and any Fe components.